In [1]:

    

'''Trains a simple convnet on the MNIST dataset.
Gets to 99.25% test accuracy after 12 epochs
(there is still a lot of margin for parameter tuning).
16 seconds per epoch on a GRID K520 GPU.
'''

import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras import backend as K

from __future__ import print_function


    

    




Using TensorFlow backend.

In [2]:

    

batch_size = 128
num_classes = 10
epochs = 12

# input image dimensions
img_rows, img_cols = 28, 28

# the data, shuffled and split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()

if K.image_data_format() == 'channels_first':
    x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
    x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
    input_shape = (1, img_rows, img_cols)
else:
    x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)
    x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)
    input_shape = (img_rows, img_cols, 1)

x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
print('x_train shape:', x_train.shape)
print(x_train.shape[0], 'train samples')
print(x_test.shape[0], 'test samples')


    

    




Downloading data from https://s3.amazonaws.com/img-datasets/mnist.npz
11476992/11490434 [============================>.] - ETA: 0sx_train shape: (60000, 28, 28, 1)
60000 train samples
10000 test samples

In [3]:

    

# convert class vectors to binary class matrices
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)

model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=input_shape))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes, activation='softmax'))

model.compile(loss=keras.losses.categorical_crossentropy, optimizer=keras.optimizers.Adadelta(), metrics=['accuracy'])

model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=1, validation_data=(x_test, y_test))
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])


    

    




Train on 60000 samples, validate on 10000 samples
Epoch 1/12
60000/60000 [==============================] - 111s - loss: 0.3353 - acc: 0.8982 - val_loss: 0.0804 - val_acc: 0.9757
Epoch 2/12
60000/60000 [==============================] - 112s - loss: 0.1123 - acc: 0.9665 - val_loss: 0.0554 - val_acc: 0.9825
Epoch 3/12
60000/60000 [==============================] - 113s - loss: 0.0841 - acc: 0.9754 - val_loss: 0.0436 - val_acc: 0.9854
Epoch 4/12
60000/60000 [==============================] - 114s - loss: 0.0706 - acc: 0.9792 - val_loss: 0.0374 - val_acc: 0.9877
Epoch 5/12
60000/60000 [==============================] - 117s - loss: 0.0630 - acc: 0.9812 - val_loss: 0.0330 - val_acc: 0.9889
Epoch 6/12
60000/60000 [==============================] - 109s - loss: 0.0557 - acc: 0.9833 - val_loss: 0.0343 - val_acc: 0.9882
Epoch 7/12
60000/60000 [==============================] - 110s - loss: 0.0506 - acc: 0.9847 - val_loss: 0.0330 - val_acc: 0.9889
Epoch 8/12
60000/60000 [==============================] - 108s - loss: 0.0454 - acc: 0.9866 - val_loss: 0.0306 - val_acc: 0.9894
Epoch 9/12
60000/60000 [==============================] - 117s - loss: 0.0434 - acc: 0.9870 - val_loss: 0.0283 - val_acc: 0.9899
Epoch 10/12
60000/60000 [==============================] - 109s - loss: 0.0403 - acc: 0.9881 - val_loss: 0.0280 - val_acc: 0.9910
Epoch 11/12
60000/60000 [==============================] - 106s - loss: 0.0391 - acc: 0.9881 - val_loss: 0.0288 - val_acc: 0.9902
Epoch 12/12
60000/60000 [==============================] - 117s - loss: 0.0367 - acc: 0.9892 - val_loss: 0.0285 - val_acc: 0.9910
Test loss: 0.0284773658265
Test accuracy: 0.991

In [ ]: